Private equity has industrialized technical diligence. The resulting system is excellent at producing reports. The more interesting question is what happens when somebody has to live with them.
The Factory
The technical diligence industry has accomplished what almost every professional-services business eventually attempts: it has turned judgment into a production system.
The economics are perfectly intelligible. Private equity produces a large number of transactions under compressed timelines. Those transactions require technical assessment. The obvious commercial response is a large bench of consultants, standardized methodologies, benchmarking databases, scoring frameworks, and enough capacity to put a team on almost any deal with very little notice.
The largest diligence firms can handle hundreds of transactions a year. Several have themselves attracted private equity ownership. CrossLake is backed by Falfurrias. Cuesta Partners sits within Riveron, which is backed by Kohlberg. There is nothing mysterious about the incentive structure. Their investors expect utilization, growth, margin, and throughput, precisely as the investors reading their reports expect from their own portfolio companies.
This model has become extremely good at manufacturing an answer to a particular question:
What does the technical condition of this company look like at the moment of acquisition?
Architecture is assessed. Security is assessed. Engineering productivity is benchmarked. Infrastructure is inspected. Risks are assigned colors, severities, scores, and recommendations. The findings are assembled into a report. The report goes to the deal team. The diligence firm proceeds to the next transaction.
For a large class of deals, this is entirely adequate.
The difficulty begins when the investment depends upon a second question:
What are you going to do about it?
That question belongs to a different economic model.
The Handoff
A diligence report enjoys one enormous advantage over a software company: the report does not have to survive contact with production.
It can recommend a platform modernization without negotiating with the engineers who must perform it. It can identify excessive infrastructure expense without discovering which architectural choices created it. It can describe an understaffed engineering organization without deciding whom to hire, whom to replace, which work to postpone, or what all of this will do to the product roadmap.
The report is allowed to end.
The company is not.
This creates the central weakness of the factory model. The people who developed the deepest understanding of the target usually disappear at precisely the moment that understanding becomes economically useful.
The diligence team works from the top down. It develops a view of the architecture, the organization, the infrastructure, the security posture, and the principal technical risks. It watches management answer difficult questions. It notices which questions produce immediate answers and which produce ten-minute excavations through Slack. It encounters undocumented dependencies, peculiar operational habits, political constraints, and the occasional indispensable engineer holding an alarming quantity of the company inside his head.
Then the transaction closes.
A new team inherits the report.
West Monroe has published research finding that more than 60 percent of PE firms incorporate diligence outputs into value-creation plans only rarely or some of the time. Their description of the problem is unusually apt. Diligence can resemble a hectic home inspection, with multiple teams identifying defects while nobody converts the collection of observations into a coherent plan for owning the house.
Source: West Monroe, “The Value Creation Missing Link in PE Due Diligence”The information survives.
The judgment that produced it does not travel nearly as well.
The Economics of Throughput
A firm performing hundreds of transactions every year cannot treat each engagement as an indefinite intellectual inquiry.
Nor should it. Such a business would collapse under its own curiosity.
A high-volume diligence practice must standardize. It must determine which questions get asked, how findings are categorized, how much senior time can be allocated, when an investigation has gone sufficiently far, and when the team must finish the report and move to the next engagement.
This produces consistency. It also imposes a boundary on diagnostic depth.
Consider the difference between discovering that infrastructure costs are high and discovering why.
On one buy-side engagement, I found that CloudFront was the target’s largest AWS expense. That is an unusual result for a CDN service and therefore an interesting one. A conventional benchmark could correctly identify that infrastructure costs were excessive. The investment question required another layer of inquiry: what architectural or configuration decision had produced the expense, whether it could be changed safely, and how much EBITDA could be recovered.
I remained with the company after close.
We found the cause and reduced that cost by 60 percent.
The distinction matters. “Infrastructure costs above benchmark” is an observation. A permanent reduction in infrastructure expense is value capture.
A scoring framework can be entirely correct and still stop several intellectual steps before the money.
The Report Problem
Private equity occasionally asks a document to perform rather heroic duties.
A technical diligence report may be expected to support the investment committee, inform negotiations, identify risks, estimate remediation effort, guide a hundred-day plan, orient the operating team, and remain useful months after the people who produced it have left.
This is a great deal to ask of a PDF.
A report is necessarily a snapshot. It records what was observed under the constraints of a transaction. It can tell the buyer that a security vulnerability exists, that the architecture has reached an awkward stage of maturity, that cloud economics are poor, or that the engineering organization cannot support the growth case in its present form.
Each of those findings creates a new set of decisions.
How urgent is the security problem? Which remediation comes first? Can the architecture be repaired incrementally? Does the investment case require a replatform? Which costs can be removed without damaging reliability? Is the engineering problem one of staffing, leadership, process, architecture, or some unpleasant combination of all four?
The report has completed its contractual duty.
Ownership has only begun its economic one.
Bain has reported that 83 percent of PE leaders believe their diligence approach has substantial room for improvement. EY has reported unexpected capability gaps discovered after close as a major problem for investors.
Sources: Bain Global PE Report 2026; EY PE Trends 2026One hardly requires a sophisticated theory to explain this.
Assessment and execution have been separated by design.
The Operator Model
Blackmere is organized around a different premise.
I lead the diligence. I stay after the transaction. I lead the execution.
The person reviewing the codebase before close is therefore aware that he may soon have to defend every recommendation to the engineers who maintain it.
The person questioning infrastructure economics may have to produce the savings.
The person calling a security vulnerability critical may be responsible for getting it out of production.
The person telling the investment committee that the engineering organization can support the underwriting may find himself sitting in the first board meeting after close with that judgment still attached to his name.
This has a useful effect on language.
It discourages ornamental findings.
It makes severity rankings expensive to exaggerate.
It forces recommendations to survive the question that every recommendation should eventually face: How, exactly?
A diligence conclusion acquires a different character when the person writing it expects to own its consequences.
Continuity Is More Than Convenience
The obvious advantage of this model is that there is no handoff.
The more important advantage is that the absence of a handoff changes the diligence itself.
Post-close implementation constraints begin informing the pre-close assessment.
A modernization recommendation has to account for the product roadmap. A cost-reduction opportunity has to account for reliability. An engineering reorganization has to account for the people already there. An AI strategy has to account for the data, workflows, economics, and existing technical organization underneath the fashionable vocabulary.
The diligence therefore becomes less interested in cataloguing imperfections and more interested in determining which imperfections matter to the investment case.
Every software company has technical debt.
Every engineering organization contains compromises.
Every sufficiently mature architecture possesses at least one decision whose original author would prefer to discuss something else.
The useful diligence question is whether any of these facts alter price, structure, operating assumptions, growth capacity, exit readiness, or the credibility of management’s account of the business.
That requires judgment.
Judgment becomes much easier to audit when the person exercising it remains present after close.
Security Makes the Difference Obvious
Security offers the clearest example because the consequences resist euphemism.
I have identified catastrophic security vulnerabilities during diligence on multiple transactions.
Under a conventional engagement, the finding enters the report. It receives an appropriate severity. The buyer closes the transaction. Someone then has to find the person who will do the remediating.
Weeks can disappear in this manner.
Under the operator model, the person who discovered the vulnerability is already present. Post-close remediation can begin immediately because discovery and ownership belong to the same engagement.
This does not make the operator clairvoyant.
It makes him accountable.
There is a considerable difference.
The Price of the Model
There is one limitation to principal-led diligence that no amount of positioning copy can abolish.
Capacity.
A person cannot personally conduct deep diligence, remain through execution, participate in management discussions, review technical evidence, design remediation, and simultaneously perform dozens of engagements.
Any operator claiming otherwise has either discovered a novel arrangement with the clock or has quietly recreated the delegation model he was supposed to replace.
Blackmere therefore carries a deliberately small number of concurrent mandates.
This is an economic sacrifice as much as a marketing position. Factory economics improve as utilization and volume rise. Operator economics eventually collide with the number of hours available to the principal.
The trade is straightforward.
Fewer transactions.
More attention per transaction.
The Evidence I Care About
The strongest evidence for the model has very little to do with the report itself.
Every diligence client I have advised to date has elected to retain me after close, including clients whose diligence concluded that no material technical remediation was required.
I do not sell post-close work during diligence.
The decision is theirs.
That retention matters because the client has already seen the work at the point when the choice is made. They have watched the diligence process, received the conclusions, and decided whether the person who developed those conclusions should remain involved in the company.
The market is perfectly capable of declining.
So far, it has chosen continuity.
Diligence team delivers a report. A different team, internal or external, translates findings into a 100-day plan. Context is lost in the transition.
The operator delivers findings, then stays to execute them. The 100-day plan is written by the person who will own it.
A scoring framework flags “infrastructure costs: above benchmark.” The report documents the risk. Root cause analysis and remediation fall to a separate team.
The operator diagnoses the root cause during diligence and eliminates it post-close. Root cause analysis and remediation are handled within a single engagement.
When the Factory Model Wins
The factory model exists for good reasons, and there are transactions where I would choose it myself.
If a lender requires a standardized technical report, a large diligence platform is an efficient solution.
If a PE firm is screening fifteen or twenty targets and wants standardized comparisons across them, a provider with thousands of historical transactions possesses benchmarking data that an independent operator cannot reproduce.
If the engagement requires technical, financial, commercial, tax, and operational diligence coordinated through one institution, a large advisory firm can assemble that machinery under one roof.
And if the deal requires technical diligence next Tuesday, capacity may decide the matter before philosophy gets a vote.
A deep bench has genuine value.
Industrialization solved genuine problems.
The mistake lies in pretending that every transaction presents the same problem.
When the Operator Model Wins
The operator model becomes more valuable as the technical condition of the company becomes more important to the investment thesis.
If the underwriting depends on platform modernization, infrastructure economics, AI adoption, security remediation, engineering improvement, or substantial technical change after close, continuity has economic value.
If the largest risk sits in execution, continuity has economic value.
If the buyer wants the person making the pre-close judgment to remain answerable for the post-close outcome, continuity has economic value.
This is especially important in software businesses where technical condition influences more than technical risk.
Architecture can constrain growth.
Infrastructure can distort gross margin.
Engineering quality can determine whether the operating case is achievable.
Security can alter the transaction itself.
Management’s account of the platform can either survive inspection or fail under it.
At that point, diligence has moved beyond inspection.
It has become part of underwriting.
Choose the Model That Matches the Decision
The diligence industry scaled by separating assessment from execution. That separation created capacity, consistency, and a very successful professional-services model.
It also created a peculiar arrangement in which the people with the freshest understanding of a software company frequently leave before anybody attempts to change it.
For transactions requiring standardized confirmation, benchmarking, rapid staffing, or broad cross-functional coverage, the factory model is well suited to the task.
For transactions where technical condition can affect the investment case and technical execution begins immediately after close, I prefer a simpler arrangement.
The person who tells you what should be done should remain long enough to discover whether he was right.